You are on page 1of 32

Data Analysis

and
Decision Making
Session 9/14 of DADM
Classroom instructions
• Please occupy chairs with your group members.

• Please download and open MS Excel data sheet sent


to you today.

• Please do not sit on the last bench. That hinders me


helping you while using SPSS.
Decision-making in Business!
VP (Sales), Pantaloon: I noticed, users search our products very much on
websites like Flipkart and Amazon.in, but I am unable to understand why
buyers in West India not buying them.

Director (Marketing): So, did you find out about buyer behaviour?

VP (Sales), Pantaloon : I need to figure out!

Director (Marketing): How are sales and profits figures in online channel
in West India?

VP (Sales), Pantaloon: I need to figure out even that too!!

Director (Marketing): I think, you should decide if you need to change


something for my online buyers.
1. Problem definition
• Management decision problem:
• Do I need to do something different for online buyers (e.g. different
marketing mix)?

• Research problem:
• What factors will affect purchase of online buyers for our products?
FOCUS GROUP
DISCUSSION
or
FOCUS GROUP
INTERVIEW
IN-DEPTH INTERVIEW

• When responses of respondents may be influenced by


fellow respondents

• Discussion of sensitive topics

• When to uncover respondents’ hidden thoughts OR


motives

• When it is not possible to gather respondents for FGD


Conceptual Framework
Product reviews

Product authenticity Purchase


satisfaction

Economic benefits
Questionnaire

• For demographic information

• For variables / factors

• For comparison of behavioural / product attributes

• For ranking OR comparison of brands


Research questions
1. Are consumer reviews more important for consumers in age group 21-30
than the consumers in age group 31-40?

2. Are consumer reviews more important for those who access online e-
commerce websites less frequently than those who access more
frequently?

3. Do consumers who buy products online less frequently consider economic


benefits more seriously than those who buy more frequently?

4. Which, among the three, is most important factor for buyers?


Conceptual Framework
Product reviews

Product authenticity Purchase


satisfaction

Economic benefits
Research Process
1. Problem definition

2. Developing the approach to the problem

3. Research design formulation

4. Data Collection

5. Data Preparation and Analysis

6. Report preparation and presentation


Research questions for Chi-square

• Does consumers’ purchase satisfaction vary across


gender?
• Does importance to Economic Benefits vary across
Gender?
• Both variables are categorical in nature
Research questions for t-tests
1. Does importance to Economic Benefits vary across
Gender?
• Independent sample t-test
2. Does Pantaloon’s Product Review (PR) score
statistically significantly vary from 4?
• One sample t-test
3. Do respondents perceive Product Reviews (PR) and
Product Authenticity (PA) differently?
• Paired sample t-test
Statistical Analysis
• Cross-tabulation and Chi-square test
• t-test
• ANOVA
• Correlation analysis
• Regression analysis
WHEN to use WHAT?
Chi-square t-test ANOVA
Nature of
First Variable Categorical Continuous Continuous
Nature of
Second Categorical Categorical Categorical
Variable
Number of
Categories 2 or more 2 More than 2
Conceptual Framework

Product reviews

Product authenticity Purchase


satisfaction

Economic benefits
Factors affecting Pantaloons’
online purchase satisfaction
Does importance to Economic Benefits vary across
Gender? Independent sample t-test

Does importance to Economic Benefits vary across age


groups? One-way ANOVA

Does importance to Economic Benefits vary across


gender and income groups? Two-way ANOVA
Factors affecting Pantaloons’
online purchase satisfaction
Does importance to Economic Benefits vary across age
groups?
Analyze  Compare Means  One-way ANOVA
EB  shift to Dependent list
Age group  shift to Factor (this is grouping factor)
Analyze  Compare Means  One-way ANOVA  Post Hoc 
Tukey
Analyze  Compare Means  One-way ANOVA  Options 
Homogeneity of variance test, Descriptive
An important
Research Question
• Are there consumers, who buy once in three
months, place more importance to Economic
Benefits than those consumers, who buy twice in a
year?
• And, if it is so, such consumers (who buy once in a
year) buy during sales only?

• Why are these questions important for Pantaloons?


Weblink for one-way
ANOVA

https
://statistics.laerd.com/spss-tutorials/one-way-anova-usin
g-spss-statistics.php
Post hoc test

 It is used to test equality of variance and subsequent
analyses
Tukey method: Advantages

 It tests all pairwise differences
 It is simple to compute
 It reduces the probability of making a Type I error
 It is also robust even for unequal group sample sizes

 unlike other methods of Post Hoc, it works even


when Test of Homogeneity of Variances is NOT
significant
Sum of Square (SS)

 It is the sum of the squares of the deviations of all
the observations, yi, from their mean. In the context
of ANOVA, this quantity is called the total sum of
squares
 Converting the sum of squares into mean squares by
dividing by the degrees of freedom lets you compare
these ratios and determine whether there is a
significant difference due to the factor. The larger
this ratio is, the more the factor will affect the
outcome.
Types of SS

 Type I: Not appropriate for factorial design
 Type II: Not appropriate for factorial design
 Type III: Appropriate for factorial design. Not
dependent on sample size, can not be used when
data has missing values
 Type IV: An extension of Type III. Not dependent on
sample size, CAN be used when data has missing
values
In output…

 Sum of square between groups: The variation due to
the interaction between the samples of two groups
 Sum of square within groups: The variation due to
differences within individual samples
 Standard error: It is the standard deviation of the
sampling distribution of a statistic
In output…

 When Leneve test of significance is not significant,
we can opt for ‘Equal variance not assumed’.
 For this, there are several post hoc procedures,
including Games-Howell, Tamhane T2, Dunnett T3
and Dunnett C.
 None of the tests are exact tests, but the Tamhane T2,
Dunnett T3 and Dunnett C are conservative
procedures.
In output…

 Levene statistics: F-test value
 df1 and df2:
 For one-way ANOVA, we have two groups that we
want to compare, so we have two cells. If we know
the mean of one of the cells and the grand mean, the
other cell must have a specific value such that (cell
mean 1 + cell mean 2) / 2 = grand mean (this
example assumes equal cell sample sizes, but
unequal cell sample sizes would not change the
number of degrees of freedom). Conclusion: for a
two-group design, df1 = 1.
In output…

 df1 and df2:
 Sticking to the one-way ANOVA, but moving on to
three groups. We now have three cells, so we have
three means and a grand mean. Again, how many
means are free to vary to get to the given grand
mean? That’s right, 2. So df1 = 2.
In output…

 The standard deviation, or SD, measures the
amount of variability or dispersion for a subject set
of data from the mean, while the standard error of
the mean, or SEM, measures how far the sample
mean of the data is likely to be from the true
population mean.
Comparison of results!
Does importance to Economic Benefits vary across
age groups?
One-way ANOVA

Does importance to Economic Benefits vary across


work status?
Comparison of results!
Does importance to Product Authenticity vary across
age groups?
One-way ANOVA

Does importance to Product Authenticity vary across


work status?
One-way ANOVA
Research Questions for t-test
and 1-way ANOVA

Product Quality
Customer
Product Benefits Loyalty

You might also like